Science fiction now
Some of Canada's leading researchers in information processing discuss the mind-boggling possibilities of their work.
by Mark Cardwell
"Future shock," Alvin Toffler wrote in his 1970 best-selling book of the same name, is "the shattering stress and disorientation that we induce in individuals by subjecting them to too much change in too short a time." That's true even in a wireless Net-driven world, where time and speed are commodities and information technologies evolve so quickly that what was considered avant-garde yesterday barely meets today's standards. To hear Canadian scientists in several sub-fields of information processing discuss the all but incredible possibilities of their work, it feels like future shock all over again.
Take Raymond Laflamme, director of the Institute for Quantum Computing at the University of Waterloo and researcher at the Perimeter Institute for Theoretical Physics. A Quebec City-born physicist, he studied under Stephen Hawking at the University of Cambridge, where he helped to change the famous scientist's mind about the reversal of the direction of time in a contracting universe (a feat the notoriously stubborn Hawking acknowledged in his own international best-seller, A Brief History of Time).
Dr. Laflamme is one of a small number of scientists around the world who are trying to harness the unpredictable nature and energy of electrons to create a new reality of information processing. "We're working with a new force of nature," says Dr. Laflamme, who holds a Canada Research Chair in quantum information. "If we're successful, it could result in a fundamental change in the basic concept or notion of information as we know it."
Information processing is generally defined as any mental or physical process used to understand and relate an observable event. As such, information processing encompasses everything from human thought and speech to writing, typing and printing texts using digital computers. Computers - machines that use numbers to perform rapid, often complex calculations by using stored instructions and information - have been considered the apex of human ability to process information since the 1950s. But the opening of new avenues like quantum information processing may soon raise the bar.
Quantum information processing (often called QIP) is a relatively new hybrid of computer science and physics. It's based on the quirky workings of quantum mechanics, which describe the nature of matter and radiation at the atomic level. Dr. Laflamme explains that conventional or "classical" computers must suppress these phenomena in order to function properly. Their main challenge is to herd electrons - the negatively charged particles of an atom that naturally fly in all directions - along wires and circuits. To do that, he says, "a Pentium processor uses a billion times more energy than it needs to perform each operation." That expended (or wasted) energy limits the ability and capacity of computers to manipulate, store and send information.
In a new twist on the old adage "if you can't lick 'em, join 'em," scientists like Dr. Laflamme have spent the past three decades theorizing, and sometimes demonstrating, how the unruly nature of electrons can be used to develop vastly more powerful ways of processing information. Using thousands or millions of electrons, conventional computers encode information on units called bits, represented by binary numbers (either 0 or 1). A calculation by a conventional computer is simply the manipulation of a string of bits. Quantum computers, however, encode their bit of information on single electrons, called qubits (pronounced Q-bits). A peculiar property of quantum mechanics called the superposition principle allows these qubits to be literally in two states, like 0 and 1, at the same time. This allows groups of electrons to perform a myriad of calculations almost instantly.
While a quantum computer might be unsuitable for everyday computer tasks like word processing or e-mail messaging, it could revolutionize fields that rely on algorithms, particularly cryptography, modeling and database indexing. "This field is still in its early stages, like computers in the 1940s, but we're already able to demonstrate that we can harness and use quantum bits," says Dr. Laflamme, who is able to coax answers to grade-school math problems from seven qubits on an experimental computer at the University of Waterloo. "There are hints of things we can perceive, but we're still puzzled and wondering about where it all may lead."
Fast-growing photonics
The same is true of the latest developments in another fast-growing subfield of information processing: photonic networks. These are progeny of fibre optics, which enable information to be carried between fixed points using light instead of electricity (making the Internet possible, among other things). Photonic networks could move beyond the fundamental limitations of electronics to provide better and more efficient processing and communication services on futuristic devices at the speed of light. "It's all very H.G. Wells-ish," concedes University of Toronto electrical and computing engineer Ted Sargent, who holds the Nortel Networks-Canada Research Chair in Emerging Technologies and is one of the country's leading researchers in the field.
For the past decade, Dr. Sargent and his group of more than a dozen researchers at U of T have been at the forefront of worldwide efforts to make photon-based networks a reality. Last summer, the Toronto team made headlines when it announced that it had made what Dr. Sargent calls "quantum dots" - polymer-coated, vibrating crystals of semi-conductors that could be linked like paperclips, forming an organic chain of molecules on which information-packed electrons could zip along in the presence of photons at speeds measured in the trillionths of seconds. "The idea would be to grow them like ivy, to crosslink them into a rich, gnarled network," says Dr. Sargent of the groundbreaking discovery, which turned what had been a theoretical concept into reality.
Many scientific hurdles remain, however, before the advent of photonic or optical networks becomes possible. One notable challenge is to build the circuits and switches needed to guide and direct photons through the network. "The hard thing about using particles of light is steering them," says Dr. Sargent. "It's like trying to herd super-fast lemmings."
If and when functioning photonic networks are developed, Dr. Sargent predicts they will change society as we know it. "Just name the field - science, medicine, economics - and the possible applications are limitless. I can see a world without cell phones, for example, where the mechanisms used for communication are almost invisible - maybe in your clothes, or your sunglasses - and would permit people to interact fully, irrespective of distance, as sensory-driven human beings, rather than on terms dictated by clunky, one-dimensional machines."
Unimaginable changes
By then, our world may have already been altered in other unimaginable ways by other new information technologies based on biology. Lila Kari, who moved to Canada from Romania in the mid-1990s, is a world-renowned researcher in the field of biological computation. She holds a Canada Research Chair in information technology and teaches computer science at the University of Western Ontario. Dr. Kari says that DNA and enzymes can, in theory, be manipulated and mixed to create a biomolecular or DNA computer that would be far superior to silicon-based systems.
Biocomputing is a broad term for scientific research aimed at determining how biology does computation, and to what extent the process can be used to create new computational models and information technologies, from the sub-cellular to the practical level. Thanks to genome research, any sequence of the four alphabet-like components - A, C, G and T - that bind together in pearl-like strings to form DNA can now be synthesized.
"Instead of encoding genetic information on DNA, which is done routinely today, why not encode other things, like numbers or letters," suggests Dr. Kari. If that could be done - and the machines, methods and nano-materials needed to make a functioning system were developed - a DNA computer would express numbers in powers of four rather than the two used in a binary system. "The sheer density of information that could be processed and stored is mind-boggling," says Dr. Kari. She estimates that "a few strands" of DNA in a solution of milk would have a storage capacity equal to 120 hectares of hard disks. "It's really a whole new order of magnitude."
The day is still far off when biocomputing will, if ever, become a reality. "There is no time frame, and I'm not sure that should be a goal," says Dr. Kari, who is organizing a DNA computing conference for 150 international experts this summer at Western. But the field offers many tantalizing possibilities. It could lead to the development of smart drugs or DNA nano-machines assigned engineering tasks such as building a nano-car and could revolutionize medicine, not to mention society. "It's like a big treasure chest," she says of the potential of biology-driven information technologies. "There are so many things you could do with them. It's funny, too, to think about how they might function. Imagine having to feed your computer ice cream instead of electricity."
In some ways, neurobiologist Naweed Syed is already doing that. Through his research at the University of Calgary into the cellular and molecular workings of neuronal mechanisms that allow people to move and breathe, the Pakistan-born physician has made groundbreaking discoveries that have helped advance international efforts to merge and harness the power of biology and nanotechnology, and to open some interesting avenues in information processing.
Nerves on a chip
Working in collaboration with the Max Planck Institute for Biochemistry in Munich, Dr. Syed's Calgary research team made headlines last year when they found that snail nerve cells cultivated on a silicon chip in the lab were able to learn, remember and transmit information to the brain. "We stimulated one nerve cell through the chip with a positive charge," relates Dr. Syed. Once excited, the cell released neurotransmitters that defused and bonded with the appropriate receptors on a second cell, which then relayed the signal to other cells in the same fashion. "The results helped us answer some fundamental questions concerning biology and neuro-electronics," says Dr. Syed. Using different techniques, similar experiments at research centres in the United States and Europe have confirmed the utility of chip technology for successful bionic hybrids. The most recent occurred in October, when researchers at the University of Florida used a group of rat brain cells on a petri dish to control the virtual flight of a fighter aircraft in a simulator.
Electronic devices that could talk directly to brains and vice versa would have wide-ranging implications, particularly in health care. According to Dr. Syed, chips could be designed and loaded with information needed to control artificial limbs, to help paralysed limbs move again and even to restore lost brain functions in people who suffer from neuropsychological disorders like Parkinson's and Alzheimer's disease. Very tiny chip particles called nano-dust could gather vital medical information from transmitters in organs and arteries throughout the body. "It would revolutionize medicine," asserts Dr. Syed.
The development of Cyborg-like hybrid computer-brain devices could offer other possibilities. Blind people could be "programmed" to see again, says the scientist. "Imagine the implications of having an electronic device inside your head that could talk to your brain - and vice versa. You'd be able to turn on your computer by just thinking about it. You'd almost become God-like.
"And remember," he goes on, "IP works both ways. As your brain became more powered by the computer, you could also make the computer feel your feelings. Science fiction is becoming reality."
